TDGM: Graph machine learning

Tamara Drucks & Franka Bause

Language: English

Credit Points: 1 CP upon agreement with the lecturers

Course description:

Graphs are everywhere and form the basis for many machine learning tasks such as content moderation or drug discovery. Unlike structured data such as images or sequences, graphs often have irregular structure and variable sizes, rendering traditional machine learning models ineffective. This workshop introduces graph machine learning and Graph Neural Networks (GNNs) from both a theoretical and practical perspective. Participants will learn how to model relational data as graphs and apply modern deep learning techniques using PyTorch Geometric. The course combines short lectures on key concepts with guided hands-on coding sessions. No prior experience with graph learning is required, but basic Python knowledge is expected.

Prerequisites:

Participants should have basic Python programming skills. Familiarity with machine learning concepts and a basic understanding of mathematics is also beneficial but not mandatory.

Biography: Tamara Drucks

I am a PhD student in the Research Unit Machine Learning at TU Wien, where I currently focus on the expressive power of graph learning algorithms and their limitations.

Biography: Franka Bause

I focus on graph learning and similarity measures for graphs, with the aim of improving efficiency, expressivity, and accuracy. I completed my doctorate with distinction in the Kriege group at University of Vienna, and am currently a Postdoc in the relational machine learning lab of Rebekka Burkholz at CISPA.